OPEN SOURCE. OPEN POSSIBILITIES.
See the flow.
Trust the result.
Turn scattered data into a story you can follow. Build SQL pipelines visually. Explore the results. Trace a number all the way home.
pip install duckpad
duckpad startPowered by DuckDB 2.0 · Local execution · MIT licensed
BUILT ON OPEN SOURCE
FROM INPUT TO INSIGHT
Less guesswork.
More “got it.”
Keep the big picture and the small details in the same place.
Think in steps.
Work in SQL.
Connect sources, joins and transformations on a visual canvas. Use guided forms or get straight to the SQL.
SELECT * FROM your_next_ideaExplore pipelines ↗02 / UNDERSTANDEvery number
has a backstory.
Click a valid result cell and follow its contributions through joins, aggregates and nested components.
Follow the lineage ↗03 / ITERATEMake a change.
Keep the context.
Save to Git, experiment on an isolated branch and compare changes visually before a reviewed merge.
Work with versions ↗A CLEAR PATH BACK
“Where did this
come from?”
Now you know.
Duckpad explains the data from the actual run. Explore the source cells behind a total, inspect the calculation, and keep following the thread.
Coverage is explicit. When a SQL construct cannot be fully explained, Duckpad tells you.
Explore lineage and its limits ↗
YOUR WORK, IN YOUR HANDS
Small setup.
Room for big ideas.
Run Duckpad on infrastructure you control. Start with local files, reuse your logic as components, and configure connections for each environment.
YOUR NEXT GOOD IDEA STARTS HERE
Let your data
make sense.
Start with Northstar, a complete example you can explore.
No external database account required.